pandas 熊猫从列中选择唯一值
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/48292656/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Pandas select unique values from column
提问by chowpay
I was able to ingest a csv in jupyter notes by doing this :
通过这样做，我能够在 jupyter 笔记中摄取 csv：
csvData= pd.read_csv("logfile.csv")
My data looks like this:
我的数据如下所示：
event_timestamp ip url
2018-01-10 00:00 111.111.111.111 http://webpage1.com
2018-01-10 00:00 222.222.222.222 http://webpage2.com
...
..
.
I got a list of unique ips:
我得到了一个独特的 ip 列表：
list_ips = csvData("[ip]")
What I'm trying to do is get a unique. Normally I would do:
我想要做的是获得一个独特的. 通常我会这样做：
list_ips.unique()
But in this case I get this error:
但在这种情况下，我收到此错误：
AttributeError: 'DataFrame' object has no attribute 'unique'
(I can use list_ips.head() and it will list a few IPs but it's not a unique list)
（我可以使用 list_ips.head() 并且它会列出一些 IP，但它不是唯一的列表）
Thanks
谢谢
EDITMy problem is I actually had:
编辑我的问题是我实际上有：
list_ips = csvData([["ip"]])
So I removed 1 set of brackets so it became:
所以我删除了一组括号，所以它变成了：
list_ips = csvData(["ip"])
Then I was able to follow Wen's example and do:
然后我就可以按照温的例子做：
list_ips.unique().tolist()
Output:
输出：
['111.111.111.111','222.222.222.222'...]
回答by YOBEN_S
You need to select the column correctly then apply unique
您需要正确选择列然后应用 unique
csvData['ip'].unique().tolist()
Out[677]: ['111.111.111.111', '222.222.222.222']
回答by Julian Rachman
The reason why you are running into this problem is because pd.read_csv("logfile.csv").unique()is not a valid attribute from DataFrame. What I suggest you do is since csvData comes out as a list, you can search for all ip's by csvData['ip']then search for unique ip's with csvData['ip'].unique().
您遇到此问题的原因pd.read_csv("logfile.csv").unique()是它不是来自 DataFrame 的有效属性。我建议你做的是因为 csvData 作为一个列表出现，你可以搜索所有 ip，csvData['ip']然后搜索带有csvData['ip'].unique().
